.[1] An empirical model for predicting low-latitude storm-time ionospheric foF2 is developed using the support vector machine technique. Considering that the ionospheric disturbances are mainly caused by interplanetary disturbances, the solar wind data are introduced as model input, as well as the ionospheric observations of Haikou (HK, with geographic coordinates of 110.3°E and 20.0°N, and geomagnetic latitudes of 8.6°N) and Chongqing (CQ, 106.5°E, 29.6°N, and geomagnetic latitudes of 18.1°N) in China. Data from 45 storms are selected as training samples to construct the model, and other 26 storms are used to validate and evaluate the model. The results indicate that the model proposed here can capture the low-latitude ionospheric disturbances most of the time. Compared with another empirical model, STORM, which has been included in International Reference Ionosphere (IRI) as storm time corrections, our model shows remarkable improvement at least for the given events.

[1] The morphology of low-latitude ionosphere is greatly affected by the zonal electric field, especially at disturbance time. In this study, historical data from two low-latitude ionosondes located at Haikou and Chongqing, China, are used to study the disturbance vertical drift properties of ionospheric F layer during the initial and main phases of 50 intense storms. The disturbance drift from an empirical model is used as an indicator to denote the E Â B disturbance magnitude. It is found that the drifts both at the base and at the peak height are increased in magnitude when the disturbance becomes stronger. For the same disturbance, ionospheric vertical drifts at the base height are comparable with those at the peak height over Haikou, but they are larger than the latter over Chongqing, both drifts being smaller than those at the magnetic equator. The drifts are larger in daytime than in nighttime, but no rule is found on their seasonal dependence. This indicates that low-latitude storm time ionospheric vertical drift of the ionospheric layer exhibits both consistent and individual properties at different locations.

[1] Using the support vector machine (SVM), an empirical local ionospheric forecasting model over Lanzhou (ELIFMOL) has been developed to predict the critical frequency of the F 2 layer ( f o F 2 ) during disturbed geomagnetic conditions. This study focuses on the reliable prediction of f o F 2 during geomagnetic storms, which is important for practical applications as well as for further understanding of the storm dynamics. In this paper, we investigate whether f o F 2 during disturbed geomagnetic conditions at a single station can be well predicted one hour ahead by using some inputs to an SVM network, such as the latest f o F 2 observations, hourly quiet time f o F 2 ( f o F 2QT ), time, and the hourly time-weighted accumulation series derived from the geomagnetic index (ap(t)). The input observation data cover the period of January 1958 to December 2000 at Lanzhou (36.1°N, 104.0°E) in China. The f o F 2 forecasted by the ELIFMOL are compared with the monthly median values and with those by the STORM MEDIAN model, the persistence model, and the STORM foF2QT model during geomagnetic storms occurring from 2001 to 2006. As for the data sets used in this paper, the results show that the performance of ELIFMOL is better than that of the latter models.

[1] This paper proposes a method for forecasting the ionospheric critical frequency, f 0 F 2 , up to 5 h ahead using the support vector machine (SVM) approach. The inputs to the SVM network are the universal time; day of the year; a 2 month running mean sunspot number (R2); a 3 day running mean of the 3 h planetary magnetic ap index, the solar zenith angle; the present value f 0 F 2 (t) and ten previously observed values f 0 F 2 (t − i), where i = 1, 2, 3, 4, 19, 20, 21, 22, 23, 24; and the six derivatives of previous 30 day running means of f 0 F 2 f m F 2 (t − j), where j = 19, 20, 21, 22, 23, 24. The output is the predicted f 0 F 2 up to 5 h ahead. The network is trained using the ionospheric sounding data from seven Chinese stations, i.e., Guangzhou, Haikou, Chongqing, Beijing, Lanzhou, Changchun, and Manzhouli stations at solar maximum and minimum. In order to test the predictive ability, the SVM was verified with different data from the training data. The quality of the proposed model prediction is evaluated by comparison with corresponding predictions from the persistence reference, the autocorrelation and the neural network (NN) models. By using data from seven Chinese stations, it is shown that the performance of the SVM model is superior to that of the autocorrelation and persistence models, and that it is comparable to that of the NN model.

No abstract